The purpose of the proposed study is to perform genetically manipulated allogeneic donor lymphocyte infusions in patients who have relapsed hematologic malignancies after allogeneic stem cell transplantation (SCT). Donor lymphocyte infusions (DL1) have resulted in the cure of some patients with relapsed leukemia or lymphoproliferative disorders after allogeneic BMT, but have been complicated by the development of graft versus host disease (GvHD). The central hypotheses of this study are the following: Retroviral mediated transfer of a chimeric CD34-Thymidine Kinase (CD34-TK75) suicide gene into human T cells to control GvHD can be performed safely, and will result in functional T cells that express a selectable marker and a suicide gene. Transduced T cells can be selected with very high efficiency using a clinical CD34+ cell selection device. After exerting anti-leukemic effects in the recipient, donor T cells can be eliminated in response to in vivo administration of ganciclovir, before damage from GvHD occurs. Initial studies from our group have shown highly efficacious killing of CD34-TK75 fusion gene-transduced murine T cells in a murine model of GvHD. These studies strongly suggest the potential for using T cell suicide gene therapy during donor lymphocyte infusions in clinical trials. The use of the CD34-TK75 fusion gene provides significant advantages over several previous studies that used bicistronic vectors, because the epitope marker gene (CD34) and the suicide gene (TK) are expressed as one protein, so there is no discordant expression resulting in inefficient selection or infusion of unmarked T cells. The expression of the chimeric CD34-TK75 suicide gene in activated T cells allows for an enhanced level of safety in the proposed clinical gene therapy trial, since cells potentially activated by retroviral insertional mutagenesis could be killed by in vivo administration of ganciclovir. The initial goal of the current application is to scale up the human T cell transduction at the Good Manufacturing Process (GMP) level, in preparation for a clinical trial (Aim 1), in the new GMP facility at Washington University. The Recombinant Advisory Committee (RAC) has approved this trial (RAC approval # 03011-566) and an IND will soon be filed with the FDA. The pre-IND meeting has been completed. The second goal is to begin enrolling patients to this trial as soon as FDA approval is granted (Aim 2). We expect that the first patient would be enrolled early in the first year of potential funding. This novel T cell suicide gene therapy/donor lymphocyte infusion study has the potential to eliminate residual leukemic cells while preventing post-infusion morbidity and mortality from GvHD. This approach could extend the life expectancy of patients with relapsed leukemia after allogeneic transplant, a population that currently has only a 50% chance of 100 day survival and only a 15-20% chance of one year survival.